Dust removal technology for driver vision leverage

• US 9,449,372 B2
• Filed: 08/12/2014
• Issued: 09/20/2016
• Est. Priority Date: 08/22/2013
• Status: Active Grant

• Current Assignee: BAE Systems Information and Electronic Systems Integration Incorporated (BAE Systems Plc)
• Original Assignee: BAE Systems Information and Electronic Systems Integration Incorporated (BAE Systems Plc)
• Inventors: Real, Edward C., Chevalier, Melissa L.
• Primary Examiner(s): SHAH, UTPAL D
• Application Number:
  US14/457,805
• Publication Number:
  US 20150055826A1
• Time in Patent Office: 770 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 2207/20182 Noise reduction or smoothing in the temporal domain; Spatio-temporal filtering
• CPC Class Codes: G06T 2207/30212 Military
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 5/10 using non-spatial domain filtering
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement
• CPC Class Codes: G06T 7/248 involving reference images or patches

Abstract

A system includes a way of improving an image obscured by airborne particles. The system includes a decomposition processor and an image generation processor. The decomposition processor decomposes an object of interest moving in a first image at a first rate of speed into at least one first subspace vector. This decomposition processor also decomposes fine particles moving at a different rate of speed than the object in the first image into at least one second subspace vector. The fine particles obscure the object of interest in a second image. The image generation processor generates based, at least in part on the first subspace vector and the second subspace vector, an image of the object without some of the fine particles obscuring the object of interest.

Claims

1. A method for improving a scene obscured by airborne particles comprising:

mathematically decomposing a first image of the scene into a first decomposition, wherein the first decomposition includes a first image subspace of an object of interest in the image having singular values and vectors of the first image;

mathematically decomposing a second image of the scene into a second decomposition, wherein the second decomposition includes a second image subspace of the object of interest having singular values and vectors of the second image that includes an airborne particle subspace associated with the airborne particles of the second image;

tracking the object of interest based, at least in part, on the first image subspace and the second image subspace to determine a location of the object in the second image; and

generating an improved image that removes at least some of the airborne particles obscuring the object in the second image, wherein the improved image is created based, at least in part, on the location of the object, the second image subspace and the airborne particle subspace by tracking changes of the singular values and vectors, wherein the generating an improved image of the object is based, at least in part, on tracked changes of the singular values and vectors.

2. The method of claim 1 wherein the mathematically decomposing the first image and the second image further comprise:

generating a singular value decomposition (SVD) of the first image; and

generating an SVD of the second image.

3. The method of claim 1 wherein the decomposition of the second image further comprises:

decomposing the second image into at least one Eigenvector.

4. The method of claim 1 wherein the decomposition of the second image further comprises:

decomposing the second image into at least one set of Eigenvector and value pairs.

5. The method of claim 4 wherein the at least one set of Eigenvector and value pairs represent at least one of the group of:

the second image subspace and the airborne particle subspace.

6. The method of claim 4 wherein the second image is represented by digital values of pixels.

7. The method of claim 6 wherein the pixel values represent voltage or a current value proportional to a power level of each pixel.

8. The method of claim 1 further comprising:

tracking the object of interest based, at least in part, on differences between the first image subspace and the second image subspace; and

generating the improved image based on differences between the first image subspace and the second image subspace.

9. The method of claim 1 wherein the method is implemented to improve an image received by an operator of a vehicle to allow the vehicle to be better maneuvered.

10. The method of claim 1 wherein the method is implemented to allow a rear view from a military vehicle such as a Mine-Resistant Ambush Protected (MRAP) military vehicle.

11. The method of claim 1 further comprising:

tracking orientations of orthogonal i, j, and k vectors as these vectors rotate from the first image to the second image; and

generating the improved image based, at least in part, on a rotational amount of the orthogonal i, j, and k vectors.

17. The method of claim 1 wherein the first and second images are back-to-back images.

12. A system for improving an image from a scene obscured by airborne particles comprising:

a decomposition processor to decompose an object of interest moving in a first image at a first rate of speed into at least one first subspace vector; and

wherein the decomposition processor is configured to decompose the airborne particles moving at a different rate of speed than the object in the first image into at least one second subspace vector, and wherein the airborne particles obscure the object of interest in a second image; and

an image generation processor configured to generate based, at least in part on the first subspace vector and the second subspace vector, an image of the object without some of the airborne particles obscuring the object of interest.

13. The system for improving an image obscured by airborne particles of claim 12 further comprising:

a subspace tracker processor configured to track singular vectors and values as the singular vector and values evolve over time.

14. The system for improving an image obscured by airborne particles of claim 13 wherein the subspace tracker further comprises:

at least one of the group of;

a projection approximation and subspace tracking algorithm, a FAST algorithm and a PASTd algorithm.

15. The system for improving an image obscured by airborne particles of claim 13 wherein the subspace tracker is applied to the sliding window of images.

16. The system for improving an image obscured by airborne particles of claim 13 wherein the subspace tracker is applied to the sliding window of a fixed number of images.